View Full Version : Hellfire Repair Log

10th May 2009, 09:19 PM
Symptom: all main sprites only being display in the ?middle? column of the display (about 4 inches wide on a 14? monitor)
Cure: Corroded track leading from pin 11 of the LS163 at 2B that goes to pin 13 of the LS157?s at locations 5J and 7J. Track repaired, fault fixed.

1st February 2010, 01:06 PM
I have just made myself a test rig, and am thinking of tackling a non working Toaplan Hellfire PCB I have. I have chosen this PCB specially because it has a few resistors & caps missing, track damage, and looks to have been bashed around. The reason I choose this is that although it may be difficult to fix, by fixing the obvious physical damage type faults I can see what difference it makes as I go along.

Firstly I want a Hellfire schematic, but I can't find one! Does anybody have one/ know where one is? I have searched a whole bunch of sites commonly referred to on AA with no luck.

Can anybody help?

1st February 2010, 01:41 PM
Probably there isnt one, very few boards have schematics available online, thankfully they are rarely needed to fix a board, unless the damage is so bad you cant see where things once went.

1st February 2010, 02:05 PM
Ahh well, I might need to refer to someone who has a working PCB to work out what values of resistors/ capacitors are missing. It is wierd, they have been either snapped off or cut off deliberately. The board is a bit of a mess, however I just want to have a play and learn as I go. Tonight I might list the locations of the missing bits for reference.

1st February 2010, 02:08 PM
manual has a list of parts inside it


[ ] Helifire.pdf 14-Jul-2002 09:46 1.6M

1st February 2010, 02:21 PM
Ah yeh, missing bits, doh!

The parts list should help, most caps outside the amp section are smoothing caps and not overly critical. Ceramic caps near TTLs are almost always 104K's

1st February 2010, 02:59 PM
Thanks Narf but I am after Hellfire (not Helifire).

I thought I found manuals/ schematics many times over but unfortunatley they were all for the latter. :(

1st February 2010, 03:05 PM
bugga :redface

1st February 2010, 09:45 PM
Well I have figured out what the missing values are, as the same ones are all over the board. They look like little resistors, but they are capacitors. The markings are Red, Red, Orange, White so I guess that makes them 2.2uF. There are in total 5 missing. Also when the board was powered up, by touching the board ever so slightly different patterns of junk came up on the screen. I can already see some track that are potentially broken, some chips with squash damage and heavy scratches, and at least one socketed chip that looks like the legs are all over the place. Plenty to do straight up just on the physical side here./

1st February 2010, 10:12 PM
I also have a damaged Hellfire board. This is one of my favourite games. Let me know if theres anything you'd like me to have a look at for you.

I have never in my life heard of capacitors marked with colour codes. Are you sure thats what they are?

2nd February 2010, 05:51 AM
I'll remove one and check to make sure, but on the PCB they have the non-polarised capacitor symbol underneath, and they are in places where there are normally capcitors. Here is a link to stuff about capacitor colour codes, scroll down to figure 3-23 http://www.tpub.com/neets/book2/3g.htm

They look like resistors, but everything else points to them being capacitors. I'll know for sure when I take one out.

2nd February 2010, 06:46 AM
They are capacitors, caps with colour codes are very common on early 90s boards, non-bootlegs mainly, they look like little resistors, about half the length and usually yellow or white with colour bands. Usually one per TTL chip where you would expect ceramic caps to be on older boards or bootlegs.

When a TTL chip switches its output it takes more current during the switch than when the gate is either high of low, this leads to a tiny dip in voltage at the chip, the caps are there to smooth out these ripples, 74LS is much lower power than original 74 series TTL so its less of an issue, but the caps are still there as its good design, leads to a stable board. If a few are missing it is not going to stop a board in its tracks, they just are insurance against noisy power supplies which may lead to unstable boards. The only time a cap is likely to halt a board is if its part of the reset delay system, ie a cap feeding a not gate on something like a 74LS04 which then connects to the Reset line on the cpu.

I wouldnt take it off if I were you, they are very fragile, it is likely to fall apart due to the heat and stress of removal.

2nd February 2010, 07:20 AM
Cool thanks Womble

2nd February 2010, 09:52 PM
When firing up the board you just get junk, but can barely make out the font of the high score table etc. One of the roms looks bad, legs bent everywhere. Pulled rom 04 off the board. 2 legs had ends snapped off, 2 legs bent under, all legs bent a bit sideways. Fired board up without that rom in. Can see more of the game now, jailbars all over screen, but game coins up, and sounds/ music are perfect! I nearly soiled myself hearing the music.
Carefully bent leg pins all back to straight positions, and repaired the legs that had broken. Dropped the rom back in the board. Now the jailbars are gone. Most sprites look perfect, some background issues, some bits don't line up properly. Also when attract screen playing, there is some random corruption on the screen, can vaguely make out some 0's in it in columns. Also slightly tapping the table the PCB is on causes colours to go missing (red, green or blue). A few more taps they come back. Very sensitive to this.

3rd February 2010, 06:57 AM
Sounds like you are almost there, check for other bad roms, oxidized sockets and any pins bent and touching on the solder side of the board. I have a rom pin straightener, a few bucks from jaycar, drop in a chip and squeeze hard, works very well.

3rd February 2010, 09:24 PM
I have now removed and replaced all socketed EPROMS. When firing up the board again, all the random corruption is gone! I still have the problem with the background graphics though. When I removed the chips, I read them all with a Wellon VP-280. When running the files through Romident, only one is recognised as being a Hellfire rom, the rest unrecognised. Logically to me this could indicate that almost all the EPROMs have data corruption, however how accurate a test is this? Is there something else I can check to confirm otherwise??

4th February 2010, 08:53 PM
Went backwards a bit tonight. Pulled the some Eproms out again, and tried to read them a couple of times then run through Romident. No luck, then I noticed that the only eprom that did come up in Romident was a different brand to all the others (which are Sharp Branded). I didn't see Sharp in the list of manufacturers to choose from in the software, will need to research this a bit more. I now just get a garbled message regarding a RAM error now - not sure if that is a good sign or not???

5th February 2010, 09:16 AM
Are you sure they are eproms? If they are mask roms you cant assume the pinout of them and they may be zapped if you try to read them as eproms.

The fact that they dont compare with the MAME set means one of a few things, either the ROMs are shot, or the game is a different version to the one MAME has, or in the case of bootleg boards that exact ROM set may not be known.

If they are eproms and you are 100% sure you are reading them correctly then I would say they are dead.

13th February 2010, 02:22 PM
I have located an area on the board where there are potentially broken tracks due to corrosion. I have fixed one track, and managed to have the game working just about 100%. However, starting the board back again I get the message "RAM Check VRAM Error FF3FFF". After a few goes I get lucky and the board will start up again. I will look at the potential broken tracks first. It is hard to know for sure becasue they go underneath an NEC D6508 which has about 100 pins, so that makes it hard to trace.

http://tehkella.net/damian/Repairs/Hellfire_Track Repair.jpg

13th February 2010, 03:21 PM
I think that all the other traces within the corroded area are fine, although it doesn't look pretty. Any time I want to see what is going on with the board it works perfectly. I bet as soon as I want to play it it will error again.

16th February 2010, 10:34 PM
Tonight after turning the PCB on and off heaps of times I often get this error "RAM Check VRAM Error XXXXXX" (where XXXXXX represents some Hex). When I eventually got the board to fire up, I got a perfect game from start to finish out of it, apart from the screen appearing to lose sync when experiencing a minor vibration which happened about 4 times. Another tap from me on the table set it right.

So obviously the board sort of fully works but is flakey. After I finished the game where I went from start to finish I removed the jamma connector, then reconnected. After that I could not get the game started again, kept getting the RAM error.

25th February 2010, 10:44 PM
I have now located the problem RAM. Firstly I suspected a HY6116AP-12 2k x 8 bit CMOS SRAM as a suspect as a couple were in areas where there had been scratched from being piled up with other PCBS. At this stage I don't have many spare bits and pieces, so I needed to obtain an equivelant RAM chip. I remember that a NES I modded has Hyundai RAM chips. So I grabbed a spare NES I have for parts, and sure enough it had 6116RAM in it. Desoldered this, and then piggy backed on each 6116 RAM chip. The RAM error is random, so I tried piggy backing a few chips until I found the one that must have been casuing the problem. I had 100% success firing up the board when piggy backing one particular RAM chip.

I will buy a socket, and drop some new RAM in the board to make the board work practically 100%.

Summary to fix this board:

Problem: graphical errors and corruption
#4 ROM had some bent and missing pins. Straightened, and repaired broken legs.
Repaired broken track as pictured a couple of posts ago.

Problem: RAM error on startup. Intermittant, does not fault all the time.
Fix:Located suspect 2k x 8bit CMOS SRAM by piggy backing known good RAM on chips until fault no longer appeared at all. Will remove faulty RAM, install socket and new RAM to permanently fix.

A side effect along the way, the ultra sensitivity of colours going missing seems to have disappeared. If this re-occurs, I am confident it will be caused by the HD74LS374 chips which trace to the Red-green-blue outputs. This chips have been badly damaged, scraped and squished sideways.

7th March 2010, 08:04 PM
Nice one. :D